Algorithms for Materialized View Design in Data Warehousing Environment

Abstract

Data warehouses may contain multiple views accessed by queries. When these views are related to each other and defined over overlapping portions of the base data, it may be more efficient not to materialize all the views, but rather to materialize certain "shared" portions of the base data, from which the warehouse views can be derived. Selecting views to materialize is one of the most important decisions in designing a data warehouse. In this paper, we present a framework for analyzing the issues in selecting views to materialize so as to achieve the best combination of good query performance and low view maintenance. We first develop a heuristic algorithm which can provide a feasible solution based on individual optimal query plans. We also map the materialized view design problem as 0-1 integer programming problem, whose solution can guarantee an optimal solution. 1 Introduction In many large organizations a number of localized databases are maintained and managed to supp...